A method and system for measuring and identifying a dented vehicle

Abstract

This invention relates to a method and system for measuring and identifying concave-bottom vehicles. The method includes the following steps: (1) monitoring and obtaining the vehicle bottom height data sequence D of the entire train of passing vehicles and the vehicle bottom height data acquisition time sequence T. D Time sequence of passing vehicles T p (2) Combine the vehicle height information with the waveform recognition algorithm to identify the concave bottom vehicle, obtain the number of concave bottom vehicle sections N, the starting index I0[n] and ending index I1[n] of the data of each concave bottom vehicle section, and the net height h of each concave bottom vehicle section. n (3) Combining the vehicle passing time information, the order information of the concave-bottomed vehicles is obtained through the order recognition algorithm; (4) Correcting the recognition results of the concave-bottomed vehicles in steps (2) and (3). This invention provides a method and system for measuring and recognizing concave-bottomed vehicles that can identify the number and clearance height of concave-bottomed vehicles in real time and match the order of concave-bottomed vehicles.

Description

Technical Field

[0001] This invention relates to the field of rail transit safety inspection technology, and in particular to a method and system for measuring and identifying concave-bottom cars in railway marshalling yard operations. Background Technology

[0002] Concave-bottom cars are a type of railway transport vehicle with a recessed floor, used to transport special goods such as large transformers, generators, and small cars. Due to their low undercarriage clearance, concave-bottom cars are prohibited from being directly shunted via hump yards during marshalling yard dismantling operations. They must be transported to prohibited shunting lines for storage, severely impacting operational efficiency. Furthermore, the undercarriage clearance height varies among different concave-bottom cars. Therefore, identifying concave-bottom cars and their order before trains arrive at the marshalling yard, accurately measuring their undercarriage clearance height in real time, and filtering out cars with excessively low clearance heights are crucial for improving the operational capacity of marshalling yards.

[0003] The current method for measuring and identifying concave-bottom vehicles is to use laser sensors to measure the clearance height of the truck's undercarriage and count the frequency of data points with heights below a certain threshold. The shortcomings of this method are: (1) inability to determine the order of concave-bottom vehicles: the order information of concave-bottom vehicles is crucial for improving dismantling efficiency; (2) limited identification types: only one type of concave-bottom vehicle can be identified (i.e., the JSQ6 type concave-bottom double-layer transport vehicle, abbreviated as "JSQ6 vehicle"); (3) poor adaptability: the identification effect is poor for trains traveling at varying speeds; (4) difficulty in error correction: it is difficult to correct misjudgments, and misjudged information will increase the workload of confirming and verifying the undercarriage height. Therefore, the current method for measuring and identifying concave-bottom vehicles still falls short of practical application requirements. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for measuring and identifying concave-bottomed vehicles that can identify the number and clearance height of concave-bottomed vehicles in real time and match their order.

[0005] This invention is achieved through the following technical solution:

[0006] A method for measuring and identifying vehicles with concave bottoms, the method comprising the following steps:

[0007] (1) Monitoring the data sequence D of the car bottom height of the entire train passing through, and the time sequence T of the car bottom height data acquisition. D Time sequence of passing vehicles T P and vehicle model information;

[0008] (2) Combining the undercarriage height information, the concave undercarriage is identified by a waveform recognition algorithm to obtain the number of concave undercarriage sections N, the starting index I0[n] and ending index I1[n] of the data for each concave undercarriage section, and the net height h of each concave undercarriage section. n, (n = 1, 2, ..., N);

[0009] (3) Combining the vehicle passing time information, the order information of the concave bottom vehicle is obtained through the order recognition algorithm;

[0010] (4) Based on the vehicle model information, correct the concave bottom vehicle identification results in steps (2) and (3).

[0011] Furthermore, in the aforementioned method for measuring and identifying concave-bottom vehicles, the waveform recognition algorithm in step (2) is as follows:

[0012] (1) The vehicle undercarriage height data sequence D and M filters F are respectively m Convolution is performed to obtain M filtered waveforms D. m ,Right now

[0013] (2) For each filtered waveform D m Find those less than the threshold δ m All D m subscript J m ;

[0014] (3) Set the subscript J of each group m Divide into N′ m Segment, traverse N′ m The start and end indices of the segments are determined. If the difference between the start index of the (i+1)th segment and the end index of the ith segment is less than the threshold δ0, then the data of the ith and (i+1)th segments are merged (i = 1, 2, ..., N′). m -1), to obtain N m The start and end indices of the data segment;

[0015] (4) Select m groups of filtered waveforms D m N m The most frequent waveform subscripts are used as the starting subscript I0[n] and ending subscript I1[n] for the data of each concave bottom car section, respectively;

[0016] (5) Select data from I0[n] to I1[n] from the vehicle bottom height sequence D, and calculate the minimum value as the net height h[n] of the nth concave bottom vehicle.

[0017] Furthermore, in the aforementioned method for measuring and identifying concave-bottom vehicles, the sequential identification algorithm in step (3) is as follows:

[0018] (1) Finding T P At each moment in T D The subscript J in P This makes T D [J P [i]]≤T P [i]≤TD [JP[i]]+1(i=1,2,...,N P );

[0019] (2) Calculate I0[n]~I1[n] and J P The overlap of the sequences is used to select the sequence with the highest overlap as the sequence L[n] of the nth concave-bottom car.

[0020] Furthermore, in the aforementioned method for measuring and identifying concave-bottom vehicles, step (4) includes:

[0021] (1) Based on the vehicle type information, obtain the vehicle type and length s[n] of each car. p ];

[0022] (2) Based on the time sequence T of the passing train P [n p ], calculate the transit time t[n] for each car. p ];

[0023] (3) Calculate the passing speed of each car.

[0024] (4) Calculate the number of data points D under each car. p [n p ]=v[n p ]*f (where f is the data sampling frequency);

[0025] (5) The number of data points D identified as concave-bottom vehicles p [n p ],like If the vehicle with the dented bottom is identified, its identification data will be retained; otherwise, if the vehicle with the dented bottom is misidentified, its data will be deleted.

[0026] (6) For vehicles with concave bottoms that are not identified in the vehicle model information, find their sequence l′ and extract the vehicle bottom height data D corresponding to the l′ vehicle. l′ Using the waveform recognition algorithm, D is found. l′ The clearance height in the middle;

[0027] (7) Update the number of concave bottom sections.

[0028] Furthermore, in the aforementioned method for measuring and identifying concave-bottomed vehicles, the vehicle types include JSQ6 vehicles, D-type vehicles, and X2H-type container flatcars.

[0029] A system for measuring and identifying concave car bottoms, comprising:

[0030] The data acquisition device is used to monitor and obtain measurement data of the entire train passing through in real time, including the car undercarriage height data sequence D and the car undercarriage height data acquisition time sequence T. D Time sequence of passing vehicles TP and vehicle model information;

[0031] The concave-bottom vehicle identification device is used to identify concave-bottom vehicles based on the measured data using a waveform recognition algorithm, obtain the number of concave-bottom vehicle sections N, and obtain the starting index I0[n] and ending index I1[n] of the data for each concave-bottom vehicle section, as well as the net height h of each concave-bottom vehicle section. n , where n = 1, 2, ..., N;

[0032] A rank recognition device is used to obtain the rank information of the concave-bottom vehicle based on the measurement data and a rank recognition algorithm.

[0033] A correction device is used to correct the identification result and ranking information of the concave-bottom vehicle based on the vehicle model information.

[0034] Furthermore, in the aforementioned concave-bottom vehicle measurement and identification system, the concave-bottom vehicle identification device employs the following waveform recognition algorithm for identification:

[0035] (1) The vehicle undercarriage height data sequence D and M filters F are respectively m Convolution is performed to obtain M filtered waveforms D. m ,Right now

[0036] (2) For each filtered waveform D m Find those less than the threshold δ m All D m subscript J m ;

[0037] (3) Set the subscript J of each group m Divide into N′ m Segment, traverse N′ m The start and end indices of the segments are determined. If the difference between the start index of the (i+1)th segment and the end index of the ith segment is less than the threshold δ0, then the data of the ith and (i+1)th segments are merged (i = 1, 2, ..., N′). m -1), to obtain N m The start and end indices of the data segment;

[0038] (4) Select m groups of filtered waveforms D m N m The most frequent waveform subscripts are used as the starting subscript I0[n] and ending subscript I1[n] for the data of each concave bottom car section, respectively;

[0039] (5) Select data from I0[n] to I1[n] from the vehicle bottom height sequence D, and calculate the minimum value as the net height h[n] of the nth concave bottom vehicle.

[0040] Furthermore, in the aforementioned concave-bottom vehicle measurement and identification system, the locating identification device obtains the locating information of the concave-bottom vehicle based on the following locating identification algorithm:

[0041] (1) Finding T P At each moment in T D The subscript J in P This makes T D [J P [i]]≤T P [i]≤T D [J P [i]]+1(i=1,2,…,N P );

[0042] (2) Calculate I0[n]~I1[n] and J P The overlap of the sequences is used to select the sequence with the highest overlap as the sequence L[n] of the nth concave-bottom car.

[0043] Furthermore, in the aforementioned concave-bottom vehicle measurement and identification system, the correction device performs the following operations:

[0044] (1) Based on the vehicle type information, obtain the vehicle type and length s[n] of each car. p ];

[0045] (2) Based on the time sequence T of the passing train P [n p ], calculate the transit time t[n] for each car. p ];

[0046] (3) Calculate the passing speed of each car.

[0047] (4) Calculate the number of data points D under each car. p [n p ]=v[n p ]*f (where f is the data sampling frequency);

[0048] (5) The number of data points D identified as concave-bottom vehicles p [n p ],like If the vehicle with the dented bottom is identified, its identification data will be retained; otherwise, if the vehicle with the dented bottom is misidentified, its data will be deleted.

[0049] (6) For vehicles with concave bottoms that are not identified in the vehicle model information, find their sequence l′ and extract the vehicle bottom height data D corresponding to the l′ vehicle. l′ Using the waveform recognition algorithm, D is found. l′ The clearance height in the middle;

[0050] (7) Update the number of concave bottom sections.

[0051] Furthermore, the concave bottom vehicle measurement and identification system includes JSQ6 vehicles, D-type vehicles, and X2H type container flatcars.

[0052] The advantages and effects of this invention are:

[0053] 1. The present invention provides a method for measuring and identifying concave-bottom vehicles, which utilizes a convolution-based waveform extraction method and multiple sets of sensors for global waveform feature extraction, directly identifying concave-bottom vehicles and their undercarriage clearance height from the vehicle undercarriage height data, thereby improving the accuracy and adaptability of the method for measuring and identifying concave-bottom vehicles.

[0054] 2. The present invention provides a method for measuring and identifying concave-bottom cars, which matches the identification results with the passing time data and car type information to identify the position of the concave-bottom car in the whole train, significantly improving the efficiency and quality of marshalling yard operations, providing convenience for on-site operations, and promoting railway freight safety.

[0055] 3. The present invention provides a method for measuring and identifying concave-bottomed vehicles, which combines preliminary identification results, vehicle type information and vehicle passage time information, and performs mutual verification and correction of the identification results, thereby achieving the optimal identification effect.

[0056] 4. The present invention provides a method for measuring and identifying concave-bottomed wagons, which can identify more and more wagon models, including but not limited to: JSQ6 wagons, D-type wagons, X2H type container flatcars and other concave-bottomed wagon models. Attached Figure Description

[0057] Figure 1 A flowchart is shown below illustrating the method for measuring and identifying concave-bottom vehicles provided by the present invention;

[0058] Figure 2 This diagram illustrates the waveform of the concave-bottom vehicle provided by the present invention and the results of algorithm matching and recognition. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be described in more detail below with reference to the accompanying drawings. The described embodiments are only some, not all, of the embodiments of this invention. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. The embodiments of this invention will be described in detail below with reference to the accompanying drawings:

[0060] Figure 1A flowchart illustrating the method for measuring and identifying concave-bottom vehicles provided by the present invention is shown. The method includes the following steps:

[0061] (1) Monitoring the data sequence D of the car bottom height of the entire train passing through, and the time sequence T of the car bottom height data acquisition. D Time sequence of passing vehicles T P And vehicle model information. Among them, T D There is a one-to-one correspondence between D and the data acquisition time, meaning that each vehicle undercarriage height corresponds to a data acquisition time, and the length of the D sequence is N. d T D The sequence length is N d T P The sequence length is N P The passing time is the moment when the rear of each car passes the undercarriage sensor. Vehicle type information is obtained by monitoring the car number information of passing vehicles, including vehicle type, length, and width.

[0062] (2) Combining the undercarriage height information, the concave undercarriage is identified using a waveform recognition algorithm to obtain the preliminary identification results: the number of concave undercarriage sections N, the starting index I0[n] and ending index I1[n] of each concave undercarriage section, and the net height h of each concave undercarriage section. n , (n = 1, 2, ..., N).

[0063] Specifically, the waveform recognition algorithm is as follows:

[0064] ① The vehicle undercarriage height data sequence D is respectively processed by M filters F m Convolution is performed to obtain M filtered waveforms D. m ,Right now Preferably, M can be 1-10.

[0065] ② For each filtered waveform D m Find those less than the threshold δ m All D m subscript J m Among them, δ m It is used for preliminary screening of the filtered waveform D m The parameters for medium-to-low altitude data are related to the calibration of the hardware acquisition system and can be selected between 140mm and 200mm.

[0066] ③ Set the subscript J of each group m Divide into N′ m Segment, traverse N′ m The start and end indices of the segments are determined. If the difference between the start index of the (i+1)th segment and the end index of the ith segment is less than the threshold δ0, then the data of the ith and (i+1)th segments are merged (i = 1, 2, ..., N′). m -1), to obtain N from the original data of the entire train passing through. mThe start and end indices of the data segment, i.e., N m This represents the initial number of concave-bottom car sections identified. Here, δ0 is a parameter used to ensure the continuity of data indices; its value is related to the acquisition frequency of the hardware system and the passing speed, and can range from 5 to 1000. For example, data points 1000 to 2000 represent the waveform of one concave-bottom car section, but in actual acquisition, the 40 data points from 1400 to 1440 may have values ​​higher than δ0 in ②. m Setting δ0 allows the algorithm to ignore these 40 outliers.

[0067] ④ Select m sets of filtered waveforms D m N m The most frequent waveform subscripts are used as the starting subscript I0[n] and ending subscript I1[n] of the data for each concave bottom car section, respectively.

[0068] ⑤ Select data from index I0[n] to I1[n] from the vehicle bottom height sequence D, and calculate the minimum value as the minimum height h[n] of the nth concave-bottom vehicle, that is, the clearance height h. n .

[0069] (3) Combine the preliminary identification results in step (2) with the vehicle passing time information, and obtain the concave bottom vehicle sequence information through the sequence identification algorithm.

[0070] Specifically, the order recognition algorithm is as follows:

[0071] ①Find T P At each moment in T D The subscript J in P This makes T D [J P [i]]≤T P [i]≤T D [J P [i]]+1(i=1,2,...,N P ).

[0072] ② Calculate I0[n]~I1[n] and J P The overlap of the sequences is used to select the sequence with the highest overlap as the sequence L[n] of the nth concave-bottom car.

[0073] (4) Based on the vehicle type information of the entire train (including the vehicle type information of each freight car), correct the identification results of the concave-bottom cars in steps (2) and (3) (including the number of concave-bottom cars N, the starting index I0[n] and ending index I1[n] of the data for each concave-bottom car, and the net height h of each concave-bottom car). n and the order L[n]).

[0074] Specifically:

[0075] ①Based on vehicle model information Info[np ], obtain the model and length s[n] of each car section. p ].

[0076] ②Based on the time sequence T of the passing train P [n p ], perform difference operation on it to obtain the transit time t[n] of each car. p ].

[0077] ③ Calculate the passing speed of each car.

[0078] ④ Calculate the number of data points D under each car section. p [n p ]=v[n p ]*f (f is the data sampling frequency).

[0079] ⑤ The number of data points D identified as concave-bottomed vehicles p [n p ],like If the car with the concave bottom is identified, its identification data is retained; otherwise, the car with the concave bottom is considered a false positive and its data is deleted. Here, δ1 and δ2 are the upper and lower limits of the ratio of the number of data points identifying the car with the concave bottom to the total number of data points for that car section, used to filter out false positives. A ratio that is too small may indicate structural issues such as the car's bogie, while a ratio that is too large indicates a false positive. The values ​​of δ1 and δ2 are also related to other factors such as equipment calibration; generally, δ1 can be taken as 0.3–0.6, and δ2 as 0.8–0.95.

[0080] ⑥ For vehicles with concave bottoms that are not identified in the vehicle model information, find their sequence l′ and extract the vehicle bottom height data D corresponding to the l′ vehicle. l′ Using the waveform recognition algorithm, D is found. l′ The lowest height in the area, i.e., the clearance height;

[0081] ⑦ Update the number of concave bottom sections and end the recognition process.

[0082] The system corresponding to the above-mentioned method for measuring and identifying cars with concave bottoms includes a data acquisition device, a car identification device, a sequence identification device, and a correction device. Specifically, the data acquisition device is used to monitor and obtain the measurement data of the entire train of cars in real time, including the car bottom height data sequence D and the car bottom height data acquisition time sequence T. D Time sequence of passing vehicles T P And vehicle model information. The concave-bottom vehicle identification device is used to identify concave-bottom vehicles based on the measurement data and through a waveform recognition algorithm, to obtain the number of concave-bottom vehicle sections N, the starting index I0[n] and ending index I1[n] of the data for each concave-bottom vehicle section, and the net height h of each concave-bottom vehicle section. nWhere n = 1, 2, ..., N. The sequence recognition device is used to obtain the sequence information of the concave-bottomed car based on the measurement data and a sequence recognition algorithm. The correction device is used to correct the recognition result and sequence information of the concave-bottomed car based on the vehicle type information of the entire train.

[0083] Specifically, the dented vehicle identification device uses the following waveform recognition algorithm for identification:

[0084] (1) The vehicle undercarriage height data sequence D and M filters F are respectively m Convolution is performed to obtain M filtered waveforms D. m ,Right now

[0085] (2) For each filtered waveform D m Find those less than the threshold δ m All D m subscript J m ;

[0086] (3) Set the subscript J of each group m Divide into N′ m Segment, traverse N′ m The start and end indices of the segments are determined. If the difference between the start index of the (i+1)th segment and the end index of the ith segment is less than the threshold δ0, then the data of the ith and (i+1)th segments are merged (i = 1, 2, ..., N′). m -1), to obtain N m The start and end indices of the data segment;

[0087] (4) Select m groups of filtered waveforms D m N m The most frequent waveform subscripts are used as the starting subscript I0[n] and ending subscript I1[n] for the data of each concave bottom car section, respectively;

[0088] (5) Select data from I0[n] to I1[n] from the vehicle bottom height sequence D, and calculate the minimum value as the net height h[n] of the nth concave bottom vehicle.

[0089] Specifically, the ranking recognition device obtains the ranking information of the concave car based on the following ranking recognition algorithm:

[0090] (1) Finding T p At each moment in T D The subscript J in P This makes T D [J P [i]≤T P [i]≤T D [J P [i]]+(i=1,2,...,N) P );

[0091] (2) Calculate I0[n]~I1[n] and J P The overlap of the sequences is used to select the sequence with the highest overlap as the sequence L[n] of the nth concave-bottom car.

[0092] Specifically, the calibration device performs the following operations:

[0093] ①Based on vehicle model information Info[n p ], obtain the model and length s[n] of each car section. p ].

[0094] ②Based on the time sequence T of the passing train P [n p ], perform difference operation on it to obtain the transit time t[n] of each car. p ].

[0095] ③ Calculate the passing speed of each car.

[0096] ④ Calculate the number of data points D under each car section. p [n p ]=v[n p ]*f (f is the data sampling frequency).

[0097] ⑤ The number of data points D identified as concave-bottomed vehicles p [n p ],like If the car with the concave bottom is identified, its identification data is retained; otherwise, the car with the concave bottom is considered a false positive and its data is deleted. Here, δ1 and δ2 are the upper and lower limits of the ratio of the number of data points identifying the car with the concave bottom to the total number of data points for that car section, used to filter out false positives. A ratio that is too small may indicate structural issues such as the car's bogie, while a ratio that is too large indicates a false positive. The values ​​of δ1 and δ2 are also related to other factors such as equipment calibration; generally, δ1 can be taken as 0.3–0.6, and δ2 as 0.8–0.95.

[0098] ⑥ For vehicles with concave bottoms that are not identified in the vehicle model information, find their sequence l′ and extract the vehicle bottom height data D corresponding to the l′ vehicle. l′ Using the waveform recognition algorithm, D is found. l′ The lowest height in the area, i.e., the clearance height;

[0099] ⑦ Update the number of concave bottom sections and end the recognition process.

[0100] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the scope of the present invention. Any equivalent changes and modifications made within the protection scope of the present invention should be considered to fall within the protection scope of the present invention.

Claims

1. A method for measuring and identifying concave-bottom vehicles, characterized in that, The method includes the following steps: (1) Monitoring and obtaining the data sequence of the car bottom height of the entire train passing through. Time series of data collection on vehicle undercarriage height Train passage time sequence and vehicle model information; (2) Combining the vehicle bottom height information, the concave-bottom vehicle is identified by the waveform recognition algorithm to obtain the number of sections of the concave-bottom vehicle. Starting index of data for each concave bottom car and termination subscript Clearance height of each concave-bottom car , ; (3) Combining the vehicle passing time information, the order information of the concave bottom vehicle is obtained through the order recognition algorithm; (4) Based on the vehicle model information, correct the concave bottom vehicle identification results in steps (2) and (3); In step (2), the waveform recognition algorithm is as follows: (1) Sequence of vehicle undercarriage height data Separately and Filters Perform convolution to obtain Strip filter waveform ,Right now , ; (2) For each filtered waveform Find those less than the threshold All subscript ; (3) Set each group of subscripts Divided into Segment, traversal The start and end indices of the segment, if the first... The starting index of the segment minus the first The segment termination index is less than the threshold. Then the first and the Segment data merging ,get The start and end indices of the data segment; (4) Select Group filter waveform middle The most frequent waveform index is used as the starting index for the data of each concave bottom car. and termination of bid ; (5) From the sequence of vehicle undercarriage height Select subscript from arrive The data is used to calculate the minimum value as the first... Clearance height of the recessed car ; In step (3), the order recognition algorithm is as follows: (1) Search Every moment in subscript , making , ; (2) Calculation ~ and The sequence overlap is used to select the position with the highest overlap as the first position. The order of the concave bottom car .

2. The method for measuring and identifying concave-bottom vehicles according to claim 1, characterized in that, Step (4) includes: (1) Based on the vehicle type information, obtain the vehicle type and length of each car. ; (2) Based on the train passing time sequence Calculate the transit time of each car. ; (3) Calculate the passing speed of each car. ; (4) Calculate the number of data points under each car. , This refers to the data sampling frequency; (5) The number of data points identified as concave-bottomed vehicles ,like If the vehicle with the concave bottom is identified, its identification data is retained; otherwise, the vehicle with the concave bottom is considered a misidentification and its data is deleted. Here, δ1 and δ2 are the upper and lower limits of the ratio of the number of data points of the vehicle with the concave bottom to the total number of data points of the vehicle. (6) For vehicles with concave bottoms that are not identified in the vehicle model information, find their priority. and extract the first Car floor height data corresponding to the section Using the waveform recognition algorithm to find The clearance height in the middle; (7) Update the number of concave bottom sections.

3. The method for measuring and identifying concave-bottom vehicles according to claim 1, characterized in that, The types of concave-bottomed wagons include JSQ6 wagons, D-type wagons, and X2H type container flatcars.

4. A system for measuring and identifying concave-bottom vehicles, characterized in that, include: The data acquisition device is used to monitor and obtain measurement data of the entire train passing through in real time, including the data sequence of car undercarriage height. Time series of data collection on vehicle undercarriage height Train passage time sequence and vehicle model information; The concave-bottom vehicle identification device is used to identify concave-bottom vehicles based on the measurement data using a waveform recognition algorithm, and to obtain the number of concave-bottom vehicle sections. To obtain the starting index of the data for each concave bottom car section and termination subscript Clearance height of each concave-bottom car ,in ; A rank recognition device is used to obtain the rank information of the concave-bottom vehicle based on the measurement data and a rank recognition algorithm. A correction device is used to correct the identification result and ranking information of the concave-bottom vehicle based on the vehicle model information; The concave-bottom vehicle identification device uses the following waveform recognition algorithm for identification: (1) Sequence of vehicle undercarriage height data Separately and Filters Perform convolution to obtain Strip filter waveform ,Right now , ; (2) For each filtered waveform Find those less than the threshold All subscript ; (3) Set each group of subscripts Divided into Segment, traversal The start and end indices of the segment, if the first... The starting index of the segment minus the first The segment termination index is less than the threshold. Then the first and the Segment data merging ,get The start and end indices of the data segment; (4) Select Group filter waveform middle The most frequent waveform index is used as the starting index for the data of each concave bottom car. and termination of bid ; (5) From the sequence of vehicle undercarriage height Select subscript from arrive The data is used to calculate the minimum value as the first... Clearance height of the recessed car ; The rank recognition device obtains the rank information of the concave-bottom vehicle based on the following rank recognition algorithm: (1) Search Every moment in subscript , making , ; (2) Calculation ~ and The sequence overlap is used to select the position with the highest overlap as the first position. The order of the concave bottom car .

5. The concave bottom vehicle measurement and identification system according to claim 4, characterized in that, The correction device performs the following operations: (1) Based on the vehicle type information, obtain the vehicle type and length of each car. ; (2) Based on the train passing time sequence Calculate the transit time of each car. ; (3) Calculate the passing speed of each car. ; (4) Calculate the number of data points under each car. , This refers to the data sampling frequency; (5) The number of data points identified as concave-bottomed vehicles ,like If the vehicle with the dented bottom is identified, its identification data will be retained; otherwise, the vehicle with the dented bottom is considered a misidentification, and its data will be deleted. (6) For vehicles with concave bottoms that are not identified in the vehicle model information, find their priority. and extract the first Car floor height data corresponding to the section Using the waveform recognition algorithm to find The clearance height in the middle; (7) Update the number of concave bottom sections.

6. The concave bottom vehicle measurement and identification system according to claim 4, characterized in that, The types of concave-bottomed wagons include JSQ6 wagons, D-type wagons, and X2H type container flatcars.

Images